joey00072 / nanoGRPO
nanoGRPO is a lightweight implementation of Group Relative Policy Optimization (GRPO)
☆97Updated this week
Alternatives and similar repositories for nanoGRPO:
Users that are interested in nanoGRPO are comparing it to the libraries listed below
- ☆115Updated last month
- ☆96Updated 9 months ago
- Code for "Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate"☆135Updated 2 months ago
- ☆137Updated 4 months ago
- ☆105Updated 2 months ago
- Benchmark and research code for the paper SWEET-RL Training Multi-Turn LLM Agents onCollaborative Reasoning Tasks☆175Updated this week
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆170Updated 3 months ago
- Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch☆164Updated 3 months ago
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning☆180Updated 3 weeks ago
- An extension of the nanoGPT repository for training small MOE models.☆123Updated last month
- A brief and partial summary of RLHF algorithms.☆127Updated last month
- minimal GRPO implementation from scratch☆72Updated last month
- A pipeline for LLM knowledge distillation☆100Updated 2 weeks ago
- ☆60Updated 11 months ago
- 🌾 OAT: A research-friendly framework for LLM online alignment, including preference learning, reinforcement learning, etc.☆320Updated this week
- Agentic Reward Modeling: Integrating Human Preferences with Verifiable Correctness Signals for Reliable Reward Systems☆84Updated last month
- Research Code for preprint "Optimizing Test-Time Compute via Meta Reinforcement Finetuning".☆90Updated last month
- Implementation of the Quiet-STAR paper (https://arxiv.org/pdf/2403.09629.pdf)☆53Updated 8 months ago
- Exploring Applications of GRPO☆177Updated this week
- Flow of Reasoning: Training LLMs for Divergent Problem Solving with Minimal Examples☆84Updated 3 weeks ago
- This is the official repository for Inheritune.☆111Updated 2 months ago
- "Improving Mathematical Reasoning with Process Supervision" by OPENAI☆108Updated last week
- Chain of Experts (CoE) enables communication between experts within Mixture-of-Experts (MoE) models☆155Updated last month
- Official repo of paper LM2☆36Updated 2 months ago
- Train your own SOTA deductive reasoning model☆83Updated last month
- Minimal hackable GRPO implementation☆206Updated 2 months ago
- ☆91Updated last month
- NeurIPS 2024 tutorial on LLM Inference☆41Updated 4 months ago
- A repository for research on medium sized language models.☆76Updated 10 months ago
- ☆77Updated 7 months ago